A Distributed Global Optimisation Environment for the European Space Agency Internal Network

Global optimisation problems arise daily in almost all operational and managerial phases of a space mission. Large computing power is often required to solve these kind of problems, together with the development of algorithms tuned to the particular problem treated. In this paper a generic dis- tributed computing environment built for the internal European Space Agency network but adapt- able to generic networks is introduced and used to distribute different global optimisation tech- niques. Differential Evolution, Particle Swarm Optimisation and Monte Carlo Method have been distributed so far and tested upon different problems to show the functionality of the environment. Support for both simple and multi-objective optimisation has been implemented, and the possibility of implementing other global optimisation techniques and integrating them into one single global optimiser has been left open. The nal aim is that of obtaining a distributed global multi-objective optimiser that is able to 'learn' and apply the best combination of the available solving strategies when tackling a generic iblack-box" problem.